Data indexing system using dynamic tags

Abstract

A data indexing system may use dynamic tags to increase findability and relativity of located datasets. The dynamic tags may be generated based on captured search log data. The system may calculate various parameters used to determine the findability and relativity of a given dataset. The system may calculate a dynamic tag score based on the calculated parameters. The system may store the dynamic tags and the associated dynamic tag score with the given dataset.

What is claimed is: 
     
       1. A method comprising:
 capturing, by a processor, search log data comprising a search term and an accessed dataset identifier; 
 determining, by the processor, a dynamic tag based at least in part on preprocessing the search term from the search log data, wherein preprocessing the search term further comprises at least one of: separating the search term into a plurality of strings, removing a leading white space, removing a trailing white space, converting each character to an uppercase character, converting each character to a lowercase character, stemming the search term, or removing a stop word from the search term; 
 calculating, by the processor, a custom term weightage (“C”) parameter representing a popularity of a dataset associated with the accessed dataset identifier, an important term weightage (“I”) parameter representing an importance of the search term in the dataset that is proportional to a frequency of appearance of the search term in the dataset, and a topic based weightage (“T”) parameter for the dynamic tag, the T parameter representing a word, a phrase, or a topic relevant to the dataset but not present in the dataset, wherein the C parameter, the I parameter, and the T parameter are calculated based at least in part on the search log data; and 
 calculating, by the processor, a dynamic tag score as a function of the C parameter, the I parameter, and the T parameter, wherein the dynamic tag score represents a relativity between the dynamic tag and the dataset, and wherein the dynamic tag score is used in an executed search to increase findability of the dataset. 
 
     
     
       2. The method of  claim 1 , further comprising calculating, by the processor, a normalized discounted cumulative gain (NDCG) score based at least in part on the dynamic tag score. 
     
     
       3. The method of  claim 2 , further comprising analyzing, by the processor, the NDCG score to determine an effectiveness of the dynamic tag in improving the executed search for the dataset. 
     
     
       4. The method of  claim 3 , further comprising filtering, by the processor, the dynamic tag based at least in part on at least one of the analyzing the NDCG score or the dynamic tag score. 
     
     
       5. The method of  claim 4 , further comprising associating, by the processor, the dynamic tag and the dynamic tag score with the dataset. 
     
     
       6. The method of  claim 1 , wherein the dataset comprises an unstructured dataset. 
     
     
       7. A system comprising:
 a processor; and 
 a tangible, non-transitory memory configured to communicate with the processor, the tangible, non-transitory memory having instructions stored thereon that, in response to execution by the processor, cause the processor to perform operations comprising:
 capturing, by the processor, search log data comprising a search term and an accessed dataset identifier; 
 determining, by the processor, a dynamic tag based at least in part on preprocessing the search term from the search log data, wherein preprocessing the search term further comprises at least one of: separating the search term into a plurality of strings, removing a leading white space, removing a trailing white space, converting each character to an uppercase character, converting each character to a lowercase character, stemming the search term, or removing a stop word from the search term; 
 calculating, by the processor, a custom term weightage (“C”) parameter representing a popularity of a dataset associated with the accessed dataset identifier, an important term weightage (“I”) parameter representing an importance of the search term in the dataset that is proportional to a frequency of appearance of the search term in the dataset, and a topic based weightage (“T”) parameter for the dynamic tag, the T parameter representing a word, a phrase, or a topic relevant to the dataset but not present in the dataset, wherein the C parameter, the I parameter, and the T parameter are calculated based at least in part on the search log data; and 
 calculating, by the processor, a dynamic tag score as a function of the C parameter, the I parameter, and the T parameter, wherein the dynamic tag score represents a relativity between the dynamic tag and the dataset, and wherein the dynamic tag score is used in an executed search to increase findability of the dataset. 
 
 
     
     
       8. The system of  claim 7 , wherein the popularity of the dataset is based at least in part on the search term. 
     
     
       9. The system of  claim 7 , wherein the C parameter is calculated based at least in part on at least one of a total number of times the dataset was accessed, a number of times the dataset was accessed based at least in part on the search term, a number of different users that accessed the dataset, a number of different datasets accessed during a search that generated the search log data, or a number of times the search term was used. 
     
     
       10. The system of  claim 7 , wherein the I parameter is calculated using a term frequency inverse document frequency (TF-IDF) model. 
     
     
       11. The system of  claim 7 , wherein the T parameter is calculated in response to the search term not being present in a search. 
     
     
       12. The system of  claim 7 , wherein the calculating the dynamic tag score further comprises weighting at least one of the C parameter, the I parameter, or the T parameter. 
     
     
       13. The system of  claim 7 , further comprising associating, by the processor, the dynamic tag and the dynamic tag score with the dataset. 
     
     
       14. A method comprising:
 receiving, by a computer-based system, a user search query comprising a search term; 
 determining, by the computer-based system, a dataset at least partially matching the search term, wherein the dataset comprises a dynamic tag, wherein the dynamic tag matches the search term; 
 calculating, by the computer-based system, a dynamic tag score for the dynamic tag as a function of a custom term weightage (“C”) parameter representing a popularity of the dataset, an important term weightage (“I”) parameter representing an importance of the search term in the dataset that is proportional to a frequency of appearance of the search term in the dataset, and a topic based weightage (“T”) parameter for the dynamic tag, the T parameter representing a word, a phrase, or a topic relevant to the dataset but not present in the dataset, wherein the dynamic tag score is used in determining the dataset to increase findability of the dataset; 
 calculating, by the computer-based system, a search score for the dataset, wherein the search score is based at least in part on the dynamic tag score associated with the dynamic tag matching the search term; and 
 generating, by the computer-based system, a search result list comprising the dataset, wherein a list order of the search result list is determined based at least in part on the search score. 
 
     
     
       15. The method of  claim 14 , further comprising capturing, by the computer-based system, search log data based at least in part on at least one of the user search query or the search result list, wherein the search log data comprises the search term and an accessed dataset identified associated with the dataset. 
     
     
       16. The method of  claim 15 , further comprising determining, by the computer-based system, the dynamic tag based at least in part on the search term. 
     
     
       17. The method of  claim 14 , further comprising calculating, by the computer-based system, a second dynamic tag score based at least in part on at least one of the C parameter, the I parameter, or the T parameter, wherein the second dynamic tag score is different from the dynamic tag score. 
     
     
       18. The method of  claim 17 , further comprising updating, by the computer-based system, the dynamic tag in the dataset by replacing the dynamic tag score with the second dynamic tag score. 
     
     
       19. The method of  claim 14 , further comprising:
 calculating, by the computer-based system, the C parameter based at least in part on at least one of a total number of times the dataset was accessed, a number of times the dataset was accessed based at least in part on the search term, a number of different users that accessed the dataset, a number of different datasets accessed during a search, or a number of times the search term was used; 
 calculating, by the computer-based system, the I parameter using a term frequency inverse document frequency (TF-IDF) model; and 
 calculating, by the computer-based system, the T parameter in response to the search term not being present in the search. 
 
     
     
       20. The method of  claim 1 , wherein preprocessing the search term further comprises applying natural language processing to the search term.